oSIST prEN IEC/IEEE 62704-4:2026

Določanje največje prostorsko povprečene specifične hitrosti absorpcije (SAR) v človeškem telesu zaradi brezžičnih komunikacijskih naprav, 100 MHz do 10 GHz - 4. del: Splošne zahteve za uporabo metode končnih elementov za izračun SAR

General Information

• Status: Not Published
• Public Enquiry End Date: 31-Aug-2026
• ICS: 13.280 - Radiation protection
• ICS: 17.220.20 - Measurement of electrical and magnetic quantities
• ICS: 33.070.01 - Mobile services in general
• Technical Committee: INIR - Electromagnetic fields in human enviornment
• Current Stage: 4020 - Public enquire (PE) (Adopted Project)
• Start Date: 23-Jun-2026
• Due Date: 10-Nov-2026
• Ref Project: prEN IEC/IEEE 62704-4:2026 - Determining the peak spatial-average specific absorption rate (SAR) in the human body from wireless communication devices, 100 kHz to 10 GHz - Part 4: General requirements for using the finite element method for SAR calculations

Standards Content (Sample)

SLOVENSKI STANDARD
01-september-2026
Določanje največje prostorsko povprečene specifične hitrosti absorpcije (SAR) v
človeškem telesu zaradi brezžičnih komunikacijskih naprav, 100 MHz do 10 GHz -
4. del: Splošne zahteve za uporabo metode končnih elementov za izračun SAR
Determining the peak spatial-average specific absorption rate (SAR) in the human body
from wireless communication devices, 100 kHz to 10 GHz - Part 4: General
requirements for using the finite element method for SAR calculations
Ta slovenski standard je istoveten z: prEN IEC/IEEE 62704-4:2026
ICS:
13.280 Varstvo pred sevanjem Radiation protection
17.220.20 Merjenje električnih in Measurement of electrical
magnetnih veličin and magnetic quantities
33.070.01 Mobilni servisi na splošno Mobile services in general
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

106/735/CDV
COMMITTEE DRAFT FOR VOTE (CDV)
PROJECT NUMBER:
IEC/IEEE 62704-4 ED2
DATE OF CIRCULATION: CLOSING DATE FOR VOTING:
2026-05-29 2026-08-21
SUPERSEDES DOCUMENTS:
IEC TC 106 : METHODS FOR THE ASSESSMENT OF ELECTRIC, MAGNETIC AND ELECTROMAGNETIC FIELDS ASSOCIATED WITH HUMAN
EXPOSURE
SECRETARIAT: SECRETARY:
Germany Mr Alexander Prokop
OF INTEREST TO THE FOLLOWING COMMITTEES: HORIZONTAL FUNCTION(S):
TC 9,TC 27,TC 29,TC 34,SC 62A,SC 62B,TC 69,TC

77,TC 78,TC 96,TC 100,TC 124,TC 125,CISPR
ASPECTS CONCERNED:
Saf ety
SUBMITTED FOR CENELEC PARALLEL VOTING NOT SUBMITTED FOR CENELEC PARALLEL VOTING
Attention IEC-CENELEC parallel voting
The attention of IEC National Committees, members of
CENELEC, is drawn to the fact that this Committee Draft
for Vote (CDV) is submitted for parallel voting.
The CENELEC members are invited to vote through the
CENELEC online voting system.
This document is still under study and subject to change. It should not be used for reference purposes.
Recipients of this document are invited to submit, with their comments, notification of any relevant patent rights of which they
are aware and to provide supporting documentation.
Recipients of this document are invited to submit, with their comments, notification of any relevant “In Some Countries”
clauses to be included should this proposal proceed. Recipients are reminded that the CDV stage is the final stage for
submitting ISC clauses. (SEE AC/22/2007 OR NEW GUIDANCE DOC).

TITLE:
Determining the peak spatial-average specific absorption rate (SAR) in the human body from wireless
communication devices, 100 kHz to 10 GHz - Part 4: General requirements for using the finite element
method for SAR calculations

electronic file, to make a copy and to print out the content for the sole purpose of preparing National Committee positions.

You may not copy or "mirror" the file or printed version of the document, or any part of it, for any other purpose without
permission in writing from IEC.

IEC/IEEE CDV 62704-4 © IEC 2026
PROPOSED STABILITY DATE: 2029
NOTE FROM TC/SC OFFICERS:
This CDV of IEC/IEEE 62704-4 Ed2 reflects the changes that have been made to the already approved IEC/IEEE
62704-1 Ed2. Therefore, no critical comments are expected and it is intended to publish IEC/IEEE 62704-4 Ed2
after the CDV is approved.
IEC/IEEE CDV 62704-4 © IEC 2026
0 CONTENTS
1 CONTENTS .2
2 FOREWORD.5
3 INTRODUCTION .8
4 1 Scope .9
5 2 Normative references .9
6 3 Terms and definitions .9
7 4 Abbreviated terms . 10
8 5 Finite element method . 10
9 5.1 Description . 10
10 5.2 Requirements on the implementation of the FEM . 11
11 6 SAR calculation and averaging . 11
12 6.1 General . 11
13 6.2 SAR averaging . 12
14 6.2.1 General . 12
15 6.2.2 Evaluation of psSAR aligned with a Cartesian mesh on an FEM mesh . 12
16 6.2.3 Homogeneous psSAR calculation . 13
17 6.2.4 Consideration of different tissue types in one averaging volume . 13
18 6.3 Power scaling . 13
19 6.4 Reporting peak spatial-average SAR and whole-body average SAR. 14
20 6.5 Referencing peak spatial-average SAR and whole-body average SAR . 14
21 7 Considerations for the uncertainty evaluation. 14
22 7.1 General . 14
23 7.2 Uncertainty due to device positioning, mesh density, and simulation
24 parameters . 15
25 7.2.1 General . 15
26 7.2.2 Mesh convergence. 15
27 7.2.3 Open boundary conditions . 16
28 7.2.4 Power budget . 17
29 7.2.5 Convergence of the aligned psSAR sampling . 17
30 7.2.6 Dielectric parameters of the phantom or body model . 17
31 7.3 Uncertainty and validation of the developed computational model of the DUT . 18
32 7.3.1 General . 18
33 7.3.2 Uncertainty of the DUT model (d ≥ λ/2 or d ≥ 200 mm). 18
34 7.3.3 Uncertainty of the DUT model (d < λ/2 and d < 200 mm) . 20
35 7.3.4 Model validation . 22
36 7.4 Uncertainty budget . 22
37 8 Code verification . 23
38 8.1 General . 23
39 8.1.1 Rationale . 23
40 8.1.2 Code performance verification . 24
41 8.1.3 Canonical benchmarks . 24
42 8.2 Code performance verification . 24
43 8.2.1 Propagation in a rectangular waveguide. 24
44 8.2.2 Planar dielectric boundaries . 28
45 8.2.3 Open boundary conditions . 30
46 8.3 Weak patch test. 31
IEC/IEEE CDV 62704-4 © IEC 2026
47 8.3.1 General . 31
48 8.3.2 Free-space weak patch test . 31
49 8.3.3 Dielectric-layer weak patch test . 36
50 8.4 Verification of the psSAR calculation . 39
51 8.4.1 Aligned psSAR averaging algorithm . 39
52 8.4.2 Homogeneous psSAR averaging algorithm . 39
53 8.5 Canonical benchmarks. 39
54 8.5.1 Mie sphere . 39
55 8.5.2 Generic dipole . 40
56 8.5.3 Microstrip line terminated with open boundary conditions . 41
57 8.5.4 psSAR calculation SAM phantom / generic phone . 41
58 8.5.5 Setup for system performance check . 42
59 Annex A (informative) Fundamentals of the finite element method . 44
60 A.1 General . 44
61 A.2 Model boundary value problem . 44
62 A.3 Galerkin weak form . 45
63 A.4 Finite element approximation . 45
64 A.5 Considerations for using FEM . 46
65 Annex B (informative) File format for field and SAR data . 47
66 Annex C (informative) Analytical solution for error calculation in weak patch-test
67 problems . 48
68 C.1 Generation of control mesh and FEM field values . 48
69 C.2 Free-space weak patch test . 48
70 C.3 Dielectric-layer weak patch test . 48
71 Annex D (normative) Supplemental files and their checksums . 51
72 Bibliography . 52
74 Figure 1 – Aligned rectangular waveguide and locations of the sample points E , E ,
01 10
75 E , E and E at which the E components are recorded . 27
11 12 21 x
76 Figure 2 – Waveguide filled half with free-space (green) and half with dielectric (blue) . 28
77 Figure 3 – Weak patch test arrangement: a free-space cube with edge length L
78 illuminated by a plane wave . 32
79 Figure 4 – Dielectric-layer weak patch test arrangement. 36
80 Figure 5 – Geometry of the microstrip line . 41
81 Figure 6 – Geometry of the setup for the system performance check according to [20] . 43
83 Table 1 – Budget of the uncertainty contributions of the computational algorithm and
84 of the rendering of the test-setup or simulation-setup. 15
85 Table 2 – Budget of the uncertainty of the developed model of the DUT . 20
86 Table 3 – Overall assessment uncertainty budget for the simulation results . 23
87 Table 4 – Results of the evaluation of the numerical dispersion characteristics to be
88 reported . 27
89 Table 5 – Results of the evaluation of the numerical reflection coefficient to be
90 reported; frequency range is indicated for each value to be reported . 30
91 Table 6 – Guiding parameters for coarse and fine mesh generation for the weak patch
92 test 33
93 Table 7 – Results of the evaluation of the error measures on the control mesh for the
94 weak patch test for the lowest order . 35
IEC/IEEE CDV 62704-4 © IEC 2026
95 Table 8 – Results of the evaluation of the error measures on the control mesh for the
96 weak patch test for the second lowest order. 35
97 Table 9 – Results of the evaluation of the error measures on the control mesh for the
98 weak patch test for the third lowest order . 35
99 Table 10 – Guiding parameters for coarse and fine mesh generation for the dielectric-
100 layered weak patch test . 37
101 Table 11 – Results of the evaluation of error measures on the control mesh for the
102 dielectric-layered weak patch test for the lowest order . 38
103 Table 12 – Results of the evaluation of error measures on the control mesh for the
104 dielectric-layered weak patch test for the second lowest order . 38
105 Table 13 – Results of the evaluation of error measures on the control mesh for the
106 dielectric-layered weak patch test for the third lowest order . 38
107 Table 14 – Results of the SAR evaluation of the Mie sphere . 40
108 Table 15 – Results of the dipole evaluation. 40
109 Table 16 – Results of the microstrip line evaluation . 41
110 Table 17 – 1 g and 10 g psSAR for the SAM phantom exposed to the generic phone for
111 1 W accepted power as specified in [19]. 42
112 Table 18 – Dielectric parameters of the setup (Table 1 of [20]) . 42
113 Table 19 – Mechanical parameters of the setup (Table 1 and Table 2 of [20]) . 43
114 Table 20 – 1 g and 10 g psSAR normalized to 1 W accepted power and feed-point
115 impedance (Table 3 and Table 4 of [20]) . 43
IEC/IEEE CDV 62704-4 © IEC 2026
117 INTERNATIONAL ELECTROTECHNICAL COMMISSION
118 ____________
120 DETERMINING THE PEAK SPATIAL-AVERAGE SPECIFIC ABSORPTION
121 RATE (SAR) IN THE HUMAN BODY FROM WIRELESS
122 COMMUNICATION DEVICES, 100 kHZ TO 10 GHZ –
124 Part 4: General requirements for using the
125 finite element method for SAR calculations
127 FOREWORD
128 1) The International Electrotechnical Commission (IEC) is a worldwide organization for standardization comprising
129 all national electrotechnical committees (IEC National Committees). The object of IEC is to promote
130 international co-operation on all questions concerning standardization in the electrical and electronic fields. To
131 this end and in addition to other activities, IEC publishes International Standards, Technical Specifications,
132 Technical Reports, Publicly Available Specifications (PAS) and Guides (hereafter referred to as "IEC
133 Publication(s)"). Their preparation is entrusted to technical committees; any IEC National Committee interested
134 in the subject dealt with may participate in this preparatory work. International, governmental and non-
135 governmental organizations liaising with the IEC also participate in this preparation.
136 IEEE Standards documents are developed within IEEE Societies and Standards Coordinating Committees of the
137 IEEE Standards Association (IEEE SA) Standards Board. IEEE develops its standards through a consensus
138 development process, which brings together volunteers representing varied viewpoints and interests to achieve
139 the final product. Volunteers are not necessarily members of IEEE and serve without compensation. While IEEE
140 administers the process and establishes rules to promote fairness in the consensus development process, IEEE
141 does not independently evaluate, test, or verify the accuracy of any of the information contained in its
142 standards. Use of IEEE Standards documents is wholly voluntary. IEEE documents are made available for use
143 subject to important notices and legal disclaimers (see http://standards.ieee.org/IPR/disclaimers.html for more
144 inf ormation).
145 IEC collaborates closely with IEEE in accordance with conditions determined by agreement between the two
146 organizations.
147 2) The formal decisions or agreements of IEC on technical matters express, as nearly as possible, an international
148 consensus of opinion on the relevant subjects since each technical committee has representation from all
149 interested IEC National Committees. The formal decisions of IEEE on technical matters, once consensus within
150 IEEE Societies and Standards Coordinating Committees has been reached, is determined by a balanced ballot
151 of materially interested parties who indicate interest in reviewing the proposed standard. Final approval of the
152 IEEE standards document is given by the IEEE Standards Association (IEEE-SA) Standards Board.
153 3) IEC/IEEE Publications have the form of recommendations for international use and are accepted by IEC
154 National Committees/IEEE Societies in that sense. While all reasonable efforts are made to ensure that the
155 technical content of IEC/IEEE Publications is accurate, IEC cannot be held responsible for the way in which
156 they are used or for any misinterpretation by any end user.
157 4) In order to promote international uniformity, IEC National Committees undertake to apply IEC Publications
158 (including IEC/IEEE Publications) transparently to the maximum extent possible in their national and regional
159 publications. Any divergence between any IEC/IEEE Publication and the corresponding national or regional
160 publication shall be clearly indicated in the latter.
161 5) IEC and IEEE do not provide any attestation of conformity. Independent certification bodies provide conformity
162 assessment services and, in some areas, access to IEC marks of conformity. IEC and IEEE are not responsible
163 for any services carried out by independent certification bodies.
164 6) All users should ensure that they have the latest edition of this publication.
165 7) No liability shall attach to IEC or IEEE or their directors, employees, servants or agents including individual
166 experts and members of its technical committees and IEC National Committees, or volunteers of IEEE Societies
167 and the Standards Coordinating Committees of the IEEE Standards Association (IEEE-SA) Standards Board,
168 for any personal injury, property damage or other damage of any nature whatsoever, whether direct or indirect,
169 or for costs (including legal fees) and expenses arising out of the publication, use of, or reliance upon, this
170 IEC/IEEE Publication or any other IEC of IEEE Publications.
171 8) Attention is drawn to the Normative references cited in this publication. Use of the referenced publications is
172 indispensable for the correct application of this publication.
IEC/IEEE CDV 62704-4 © IEC 2026
173 9) Attention is drawn to the possibility that some of the elements of this IEC/IEEE Publication may require use of
174 material covered by patent rights. By publication of this standard, no position is taken with respect to the
175 existence or validity of any patent rights in connection therewith. IEC or IEEE shall not be held responsible for
176 identifying Essential Patent Claims for which a license may be required, for conducting inquiries into the legal
177 validity or scope of Patent Claims or determining whether any licensing terms or conditions provided in
178 connection with submission of a Letter of Assurance, if any, or in any licensing agreements are reasonable or
179 non-discriminatory. Users of this standard are expressly advised that determination of the validity of any patent
180 rights, and the risk of infringement of such rights, is entirely their own responsibility.
181 International Standard IEC/IEEE 62704-4 has been prepared by IEC technical committee
182 TC 106: Methods for the assessment of electric, magnetic and electromagnetic fields
183 associated with human exposure, in cooperation with International Committee on
184 Electromagnetic Safety of the IEEE Standards Association, under the IEC/IEEE Dual Logo
185 Agreement.
186 This publication is published as an IEC/IEEE Dual Logo standard.
187 This edition includes the following significant technical changes with respect to the previous
188 edition:
189 a) specification of a SAR averaging algorithm for the calculation of the psSAR from
190 computational results for arbitrarily shaped homogeneous phantoms, which avoids
191 inconsistencies with respect to the algorithm for SAR averaging algorithm specified in
192 IEC/IEEE 62209-1528:2020 for measurement methods;
193 b) provision of SHA-256 checksums for all supplemental files.
194 This standard contains attached files in the form of CAD models and reference results
195 described in Annex B. Download links and checksums for these files can be found in Annex D.
196 The text of this standard is based on the following IEC documents:
FDIS Report on voting
106/###/FDIS 106/###/RVD
198 Full information on the voting for the approval of this standard can be found in the report on
199 voting indicated in the above table.
200 International Standards are drafted in accordance with the rules given in the ISO/IEC
201 Directives, Part 2.
202 A list of all parts in the IEC/IEEE 62704 series, published under the general title Determining
203 the peak spatial-average specific absorption rate (SAR) in the human body from wireless
204 communications devices, 100 kHz to 10 GHz, can be found on the IEC website.
205 Future documents in this series will carry the new general title as listed in the preceding
206 paragraph. Titles of existing documents in this series will be updated at the time of the next
207 edition.
208 The IEC technical committee and IEEE technical committee have decided that the contents of
209 this publication will remain unchanged until the stability date indicated on the IEC web site
210 under "http://webstore.iec.ch" in the data related to the specific publication. At this date, the
211 publication will be
212 • reconfirmed,
213 • withdrawn,
214 • replaced by a revised edition, or
215 • amended.
IEC/IEEE CDV 62704-4 © IEC 2026
IMPORTANT – The 'colour inside' logo on the cover page of this publication indicates that it
contains colours which are considered to be useful for the correct understanding of its
contents. Users should therefore print this document using a colour printer.
IEC/IEEE CDV 62704-4 © IEC 2026
218 INTRODUCTION
219 Finite element methods have reached a level of maturity that allows their application in
220 specific absorption rate (SAR) assessment of professional-use and consumer-use wireless
221 communication devices. In the recent past, SAR compliance assessment for small
222 transmitters was performed almost exclusively using measurements. Some wireless
223 communication devices are used in situations where laboratory SAR assessment is extremely
224 complex or not possible at all. The benefits of consensus standards to users and regulators
225 include standardized and accepted protocols, verification and validation techniques,
226 benchmark data, reporting format and means for estimating the overall assessment
227 uncertainty in order to produce valid, repeatable, and reproducible data.
228 The purpose of this document is to specify computational techniques and models to determine
229 peak spatial-average specific absorption rates (SAR). SAR will be determined by applying
230 finite element method computations of the electromagnetic field conditions produced by
231 wireless communication devices in models of the human anatomy. Intended users of this
232 document are (but are not limited to) wireless communication device manufacturers, service
233 providers for wireless communication that are required to certify that their products comply
234 with the applicable SAR limits, and government agencies.
235 Several methods described in this document are based on techniques specified in
236 IEC/IEEE 62704-1:2025.
IEC/IEEE CDV 62704-4 © IEC 2026
237 DETERMINING THE PEAK SPATIAL-AVERAGE SPECIFIC ABSORPTION
238 RATE (SAR) IN THE HUMAN BODY FROM WIRELESS
239 COMMUNICATION DEVICES, 100 kHZ TO 10 GHZ –
241 Part 4: General requirements for using the
242 finite element method for SAR calculations
244 1 Scope
245 This part of IEC/IEEE 62704 describes the concepts, techniques, and limitations of the finite
246 element method (FEM) and specifies models and procedures for verification, validation and
247 uncertainty assessment for the FEM when used for determining the peak spatial-average
248 specific absorption rate (psSAR) in phantoms or anatomical models. It recommends and
249 provides guidance on the modelling of wireless communication devices, and provides
250 benchmark data for simulating the SAR in such phantoms or models.
251 This document does not recommend specific SAR limits because these are found elsewhere
252 (e.g. in IEEE Std C95.1 [1] or in the guidelines published by the International Commission on
253 Non-Ionizing Radiation Protection (ICNIRP) [2]).
254 2 Normative references
255 The following documents are referred to in the text in such a way that some or all of their
256 content constitutes requirements of this document. For dated references, only the edition
257 cited applies. For undated references, the latest edition of the referenced document (including
258 any amendments) applies.
259 IEC/IEEE 62209-1528:2020, Measurement Procedure for the Assessment of Specific
260 Absorption Rate of Human Exposure to Radio Frequency Fields From Hand-Held and Body-
261 Mounted Wireless Communication Devices: Human models, instrumentation, and procedures
262 (Frequency range of 4 MHz to 10 GHz)
263 IEC/IEEE 62704-1:2025, Determining the peak spatial-average specific absorption rate (SAR)
264 in the human body from wireless communications devices, 100 kHz to 10 GHz – Part 1:
265 General requirements for using the finite-difference time-domain (FDTD) method for SAR
266 calculations
267 3 Terms and definitions
268 For the purposes of this document, the following terms and definitions apply.
269 ISO, IEC, and IEEE maintain terminological databases for use in standardization at the
270 following addresses:
271 • IEC Electropedia: available at https://www.electropedia.org/
272 • ISO Online browsing platform: available at https://www.iso.org/obp
273 • IEEE Dictionary Online: available at
274 https://ieeexplore.ieee.org/browse/standards/dictionary
___________
Numbers in square brackets refer to the Bibliography.
IEC/IEEE CDV 62704-4 © IEC 2026
3.1
276 mesh
277 discrete representation of the simulation model as a
278 set of voxels in a regular three-dimensional Cartesian arrangement
279 Note 1 to entry: In the scientific literature, the mesh is often referred to as a "grid."
280 [SOURCE: IEC/IEEE 62704-1:2025, 3.17, modified – The specific context " 281 time-domain method>" has been added.]
282 3.2
283 mesh
284 discrete representation of the simulation model as a set of
285 tetrahedral elements in an irregularly three-dimensional arrangement
286 Note 1 to entry: In the scientific literature, the mesh is often referred to as a "grid."
287 3.3
288 element
289 smallest three-dimensional part of a mesh
290 EXAMPLE A voxel or a tetrahedron.
291 3.4
292 subregion
293 spatially limited three-dimensional region within a computational domain
294 4 Abbreviated terms
295 ASCII American Standard Code for Information Interchange
296 BVP boundary value problem
297 DoF degrees of freedom
298 DUT device under test
299 FDTD finite-difference time-domain
300 FEM finite element method
301 PDE partial differential equation
302 PEC perfect electric conductor
303 PMC perfect magnetic conductor
304 psSAR peak spatial-average specific absorption rate
305 SAR specific absorption rate
306 SI international system of units
307 TVFE tangential vector finite elements
308 5 Finite element method
309 5.1 Description
310 This document describes applications of the finite element method (FEM) to calculate the
311 specific absorption rate (SAR). Reasons for using FEM include its proven track record in a
312 broad range of electromagnetic applications, and its ability to use an unstructured, usually
313 tetrahedral, mesh that conforms to complicated geometries, employing arbitrarily small
314 elements where needed and larger elements elsewhere.
315 Multiple ways exist to solve Maxwell’s equations with FEM. Implementations can be based on
316 field quantities or on potential quantities, and may be formulated using either the weighted
IEC/IEEE CDV 62704-4 © IEC 2026
317 residual method or the variational method [3], [4]. The weighted residual method starts
318 directly from the partial differential equation (PDE) of the boundary value problem, whereas
319 the variational method starts from the variational representation of the boundary value
320 problem. All implementations have the following in common:
321 a) They are based on PDEs, not on integral equations. The PDEs are derived from Maxwell’s
322 equations augmented by proper boundary conditions in order to frame a well-defined
323 boundary value problem on a finite computational domain.
324 b) The size of the computational domain is finite. Radiation towards infinity is implemented
325 through an open boundary condition on its outer boundaries. Radiated fields outside the
326 domain can be computed by integrating over a boundary that encloses the radiating
327 structure.
328 c) After applying excitations and boundary conditions and discretizing the computational
329 domain into a mesh, the derived PDE is transformed into a matrix equation in which the
330 matrix is large, sparse, and banded. "Large" is a consequence of having a large number of
331 unknowns, several per element on a large mesh. "Sparse" and "banded" are
332 consequences of the fact that all interactions are formulated as local interactions.
333 d) In the limit of infinitesimally small elements, the solution approaches the exact solution of
334 the PDE.
335 Annex A contains more information on FEM, along with references to literature and a
336 discussion of its limitations. Clause 8 describes a set of tests is described that shall be used
337 to verify whether a particular implementation of FEM is correct and sufficiently accurate to be
338 used for SAR calculations.
339 This document refers to Nédélec elements of the first kind, which are polynomially exact up to
340 order 0 (H (curl) or edge elements) as lowest order; up to order 1 (H (curl) elements) as
0 1
341 second lowest order; and up to order 2 (H (curl) elements) as third lowest order [5]. If an
342 implementation of the FEM is applied with one of these orders, the respective parts of the
343 code verification shall be executed with this order.
344 5.2 Requirements on the implementation of the FEM
345 The implementation of the FEM shall follow the definition in 5.1. Moreover, it shall implement
346 SAR calculation and averaging according to Clause 6. The implementation of the FEM shall
347 be verified according to Clause 8.
348 6 SAR calculation and averaging
349 6.1 General
350 The local specific absorption rate (SAR) in a location in tissue is given in Formula (1):
σE
351  (1)
SAR =
2ρ
352 where ρ is the mass density of the tissue, E is the magnitude of the electric field vector, and σ
353 is the electric conductivity. Since the local SAR can vary strongly with position, the quantity of
354 interest is often the peak spatial-average SAR averaged over a cubical volume.
355 NOTE The spatial-average SAR (sSAR) is averaged over a specified mass with a specified volume, e.g. 1 g or
356 10 g of tissue in the shape of a cube as required by FCC and [1], [2].
IEC/IEEE CDV 62704-4 © IEC 2026
357 6.2 SAR averaging
358 6.2.1 General
359 The objective of the methods to evaluate the psSAR described here is to yield results that
360 correspond to the methods and definitions of Subclause 6.3 of IEC/IEEE 62704-1:2025, which
361 describes the computation of the psSAR a) in cubical volumes aligned with a rectangular
362 mesh (6.2.2) and b) in cubical volumes following the orientation of the surfaces of arbitrarily
363 shaped phantoms filled with homogeneous tissue simulant (6.2.3). Depending on the phantom
364 or body model or on the application, one of these two algorithms shall be applied to calculate
365 the psSAR using FEM simulations within this document.
366 NOTE An example of a phantom filled with homogeneous tissue simulant is the SAM phantom specified in
367 IEC/IEEE 62209-1528:2020. Its surface follows the shape of a human head.
368 6.2.2 Evaluation of psSAR aligned with a Cartesian mesh on an FEM mesh
369 6.2.2.1 General
370 Since the algorithm for the computation of the psSAR in cubical volumes aligned with the
371 mesh in IEC/IEEE 62704-1:2025 is specified for rectilinear meshes with varying mesh step,
372 the vector components of the electric fields, the conductivity, and the mass density of the
373 finite element mesh shall be resampled on a Cartesian mesh. The resampling is carried out
374 with increasingly fine mesh steps until convergence of the dissipated power is reached in the
375 regions where local SAR maxima are located. In order to reduce the computation time for the
376 iterative resampling and SAR averaging, subregions with local SAR maxima are identified in a
377 pre-scan. In these subregions, the psSAR is then calculated according to Subclauses 6.3.1
378 (aligned) or 6.3.2 (homogeneous) of IEC/IEEE 62704-1:2025. The maximum psSAR of all
379 subregions shall be reported as the psSAR maximum together with its interpolation
380 uncertainty.
381 The following steps shall be carried out to resample the geometry and the power density in a
382 set of regions around local SAR maxima for the application of the SAR averaging algorithms
383 of IEC/IEEE 62704-1:2025.
384 a) Specify an orientation of a rectilinear mesh relative to the coordinate system of the FEM
385 mesh considering the relevant features of the model; this orientation shall align with
386 surface planes or conducting planes of the phantom or of the DUT.
387 b) Iteratively resample the geometry and local SAR distribution in the rectilinear mesh and
388 evaluate psSAR at each iteration until convergence is achieved (see 6.2.2.2).
389 c) Report the highest psSAR of all subregions together with its interpolation uncertainty.
390 6.2.2.2 Calculation of the aligned psSAR on an iteratively refined rectangular mesh
391 The aligned psSAR according to 6.3.1 IEC/IEEE 62704-1:2025 shall be evaluated on a
392 rectilinear mesh that encompasses a region around a local SAR maximum with individual
393 equidistant mesh steps for each axis. Each mesh cell is assigned the local distribution of the
394 dissipated power, the conductivity, and the mass density.
395 a) The mass density for each mesh cell shall be assigned by nearest-neighbour interpolation
396 of the mass density distribution of the tetrahedral mesh.
397 b) The conductivity for each mesh cell shall be assigned by nearest-neighbour interpolation
398 of the mass density distribution of the tetrahedral mesh.
399 c) In the mesh cells that have a mass density different from zero, the dissipated power
400 density is calculated by evaluating the electric field of the finite element mesh in the
401 centre of the mesh cell of the rectilinear mesh.
402 d) The initial mesh step length Δ for each axis of the rectilinear mesh shall be calculated in
403 accordance with Formula (2):
IEC/IEEE CDV 62704-4 © IEC 2026
m
404  (2)
Δ ≤ 3
ρ
max
405 where
406 m is the averaging mass of the target volume;
407 ρ is the maximum mass density of the geometry in the computational domain.
max
408 e) The psSAR for the region under evaluation shall be calculated on the initial mesh
409 according to the procedure specified in Subclause 6.3.1 of IEC/IEEE 62704-1:2025. Then
410 the region shall be resampled on a rectilinear mesh with a reduced mesh step size Δ =
i+1
411 0,5 Δ . This procedure shall be repeated until the difference in psSAR from the previous
i
412 iteration to the present iteration is less than 1 %.
413 6.2.3 Homogeneous psSAR calculation
414 The specification of the psSAR averaging algorithm of Subclause 6.3.2 of IEC/IEEE
415 62704-1:2025 does not depend on the topology of the mesh applied for the computational
416 evaluation of the field distribution in the phantom. It shall therefore be applied on FEM
417 simulations as specified.
418 6.2.4 Consideration of different tissue types in one averaging volume
419 When averaging SAR over tissue in the head or trunk, or over an extremity or limb, according
420 to 6.2.2, the guidance of Subclause 6.3.3 of IEC/IEEE 62704-1:2025 applies.
421 6.3 Power scaling
422 In FEM simulations, the accepted power is generally delivered to the device by means of a
423 port with known characteristic impedance. Depending on the input impedance of the device, a
424 specific power level is accepted by the antenna or load. The simulation results, including SAR,
425 will be relative to this accepted power. To obtain the SAR for a different accepted power level,
426 such as the target accepted power, the SAR results shall be adjusted by scaling using
427 Formula (3):
P
acc,target
428  (3)
SAR = SAR
scaled scaled
P
acc,computed
429 where
430 P is the target accepted power;
acc,target
431 P is the accepted power computed by the FEM simulation.
acc,computed
432 P is the power delivered to the load by the simulation, which is found from the
acc,computed
433 complex voltage and current at the feed-point of the FEM mesh in accordance with Formula
434 (4):
∗
435  (4)
P = Re UI
{ }
acc,computed
436 where U and I are complex quantities, and the asterisk indicates the complex conjugate.
437 If an incident plane wave source is applied, SAR can be scaled based on the incident power
438 density. The incident power density can be computed using Formula (5):
IEC/IEEE CDV 62704-4 © IEC 2026
∗
439  (5)
P Re EH×
inc { inc inc}
440 where E and H represent the incident electric field and magnetic field from the plane
inc inc
441 wave. The computed incident power density can then be used to scale the SAR in the same
442 manner as the accepted power.
443 Changes in SAR due to performance variations in radio frequency (RF) components that
444 affect P (due to thermal, electrical, or other tolerances) shall be determined during
acc,target
445 laboratory validation of the computational model of the DUT (see 7.3). It shall be considered
446 either by choosing the maximum possible value for P or by adding the performance
acc,target
447 variation in the uncertainty budget (see 7.4).
448 6.4 Reporting peak spatial-average SAR and whole-body average SAR
449 The peak spatial-average SAR shall be reported. If applicable, separate values for both body
450 and extremity tissues shall be reported (see 6.2.4). In addition, the coordinates of two corner
451 points of the averaging cube shall be reported. The first corner point is defined by the
452 minimum x-, y- and z-coordinates of all six corner points of the cube, the second corner point
453 by its maximum x-, y- and z-coordinates.
454 The selected averaging algorithm shall be reported: aligned psSAR calculation (6.2.3) or
455 homogeneous psSAR calculation (6.2.4). For homogenous psSAR calculation, the reference
456 coordinate system with respect to the phantom shall be reported (6.3.2, Step a) of IEC/IEEE
457 62704-1:2025).
458 The psSAR and whole-body average SAR shall be reported with their computational
459 uncertainty.
460 6.5 Referencing peak spatial-average SAR and whole-body average SAR
461 Test results that are identified as compliant with the peak spatial-average SAR and whole-
462 body average SAR procedures of IEC/IEEE 62704-4 shall satisfy the definition of the
463 description of the FEM of Clause 5 and the SAR averaging requirements of Clause 6.
464 Moreover, the simulation software shall have passed the code verification of Clause 0. The
465 peak spatial-average SAR and the whole-body average SAR shall be referenced both, in
466 written publications and in the computational software as “Peak Spatial-Average SAR
467 according to IEC/IEEE 62704-4, Ed. 2” or “psSAR according to IEC/IEEE 62704-4, Ed. 2”.
468 The selected averaging algorithm shall be reported: aligned psSAR calculation (6.2.2) or
469 homogeneous psSAR calculation (6.2.3).
470 7 Considerations for the uncertainty evaluation
471 7.1 General
472 Assuming the FEM code has been implemented correctly, which shall be determined with the
473 tests described in Clause 8, some uncertainties remain. This Clause 7 shows how they shall
474 be evaluated to obtain a measure of overall assessment uncertainty. It follows the
475 computational uncertainty scheme of Clause 7 in IEC/IEEE 62704-1:2025, with modifications
476 appropriate to FEM. As stated in the cited clause, the computational uncertainties are divided
477 into the following three categories:
478 a) discretization accuracy and uncertainty due to mesh density,
479 b) computational accuracy of the specific FEM implementation,
480 c) accuracy of the computational representation of the actual DUT.
=
...